Node >v16.6.2 results in ECONNRESET behind proxy

[erikbrgn]

Version

=16.7.0

Platform

Darwin 21.5.0 Darwin Kernel Version 21.5.0: Tue Apr 26 21:08:37 PDT 2022; root:xnu-8020.121.3~4/RELEASE_ARM64_T6000 arm64

Subsystem

No response

What steps will reproduce the bug?

Hard to say as it might be connected to company proxy/firewall, however, running (any) npm command results in a ECONNRESET error on node versions above 16.6.2 (tried every minor version above up until LTS 18). Everything below 16.7.0 works fine.

This has to be done behind a proxy, set npm proxy by npm config set proxy http://YOURPROXY and npm config set https-proxy http://YOURPROXY (if there is one). Also make sure to set appropriate env variables.

Use nvm to easily switch between node versions and make sure to clean cache.

How often does it reproduce? Is there a required condition?

Everytime.

What is the expected behavior?

No ECONNRESET error on node versions higher than 16.6.2

What do you see instead?

npm ERR! code ECONNRESET
npm ERR! syscall read
npm ERR! errno -54
npm ERR! network read ECONNRESET
npm ERR! network This is a problem related to network connectivity.
npm ERR! network In most cases you are behind a proxy or have bad network settings.
npm ERR! network
npm ERR! network If you are behind a proxy, please make sure that the
npm ERR! network 'proxy' config is set properly. See: 'npm help config'

Additional information

I've tried on a different machine with the same results (Linux 5.4.0-121-generic #137-Ubuntu SMP Wed Jun 15 13:33:07 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux.). I can't seem to find anything between 16.6.2 and 16.7.0 that might be responsible for this behavior though. Perhaps someone here can enlighten me.

[bnoordhuis]

ECONNRESET means the other end closed the connection. You're reporting this as a bug, which it probably isn't, at least not in node. I can convert it to a discussion if help is more what you're looking for.

[erikbrgn]

Sure.

I was thinking something must’ve been introduced in 16.7.0 that changed the behavior (perhaps an unintended bug), but in the end help (be it a discussion) would be great.

[bnoordhuis]

If you think it's a regression (could be, it happens) and are in a position to run git-bisect, that'd be very helpful. I went over the list of commits between v16.6.2 and v16.7.0 and nothing obvious stands out.